Machinability Research on the Micro-Milling for Graphene Nano-Flakes Reinforced Aluminum Alloy

Round 1

Reviewer 1 Report

The manuscript "Machinability Research on the Micro-milling for Graphene Nano-flakes Reinforced Aluminum Alloy" reports on the fabrication of graphene reinforced aluminum alloy composites via powder metallurgy approach and on the study of the effect of varying graphene nano-flakes content on the machinability of composites.

From a general point of view, the topic of the manuscript is interesting and worth of investigation. Even if such topic is already studied by some years, the authors are able in reporting some new and interesting approaches and results.

The introduction clearly states the aim of the work and inserts the work within a general coherent technological framework. The experimental section is complete and detailed. The experimental approaches are reliable and well-founded. 

The section devoted to experimental results and discussions is generally founded. However, it appears as a succession of experimental results without strong discussions on the microscopic mechanisms and processes justifying the results.

So, even if the manuscript is generally interesting, it requires some important improvements before publication:

1) The main missing part concerns a discussion on the microscopic aspects related to metal-graphene interaction making this hybrid composite so interesting. The authors should outline, from an atomic point of view, how the synergistic combination of the properties of metals and graphene leads to improved properties focusing on the basic aspect (energetics, etc.). See, as examples: International Material Reviews 62, 241 (2017); Crystals 7, 219 (2017); Chem. Soc. Rev. 41, 666 (2012).

In addition, the authors, when discussing the experimental data and results, should strongly highlight what are the microscopic parameters of Al and graphene justifying the results and composite behaviour. As examples:

a) what is the reason for the “rise-fall-rise” trend of cutting force in Fig. 6? The reported discussion seems too qualitative.

b) Fig. 7: when you state "This is mainly attributed to the two factors: grain refinement effect of graphene and agglomeration of graphene.": why?

c) "This is attributed to two factors: on the one hand, the wrinkle structure of graphene can inhibit the coarsening and growth of aluminum matrix grains, which resulting the finer grain and hence improvement of composites plasticity. On the other hand, since graphene is carbon particle in nature, graphene as reinforcing phase in composites have a certain lubrication effect also, thus improving the friction coefficient between tool and workpiece, and reducing the generation of friction force.". These seem speculations if quantitative data are not reported.

and so on.

2) Table 1: for the experimental derived parameters, corresponding errors should be reported.

3) Fig. 9 and surface roughness: surface roughness is a very important parameter characterizing the surface morphology, topography and topology (see: A.-L. Barabasi, H.E. Stanley, Fractal Concepts in Surface
Growth (Cambridge University Press, Cambridge, 1995); Nanoscale Research Letters 4, 262 (2009); Journal of Materials Processing Technology 123, 133 (2002)). Some discussions in this sense could be very useful.

4) Some Raman and X-ray diffraction measurement could be useful (at least as perspectives) in characterize the graphene flakes quality and the graphene-Al interaction.

Author Response

Thank you for giving us the opportunity to revise our manuscript and for the thoughtful and valuable comments on this manuscript. We have checked through the whole paper and corrected errors. At same time, we add Raman, coefficient of friction and grain size data in the revised paper. All the changes are highlighted in the Marked Revision. The point-to-point responses to the comments are listed as following.

Point 1: The main missing part concerns a discussion on the microscopic aspects related to metal-graphene interaction making this hybrid composite so interesting. The authors should outline, from an atomic point of view, how the synergistic combination of the properties of metals and graphene leads to improved properties focusing on the basic aspect (energetics, etc.). See, as examples: International Material Reviews 62, 241 (2017); Crystals 7, 219 (2017); Chem. Soc. Rev. 41, 666 (2012).

Response 1: In the revised manuscript, we added Raman, wear resistance and grain size data to illustrate the metal-graphene interaction. However, due to our research limitation, we have not yet been able to explain the interaction between graphene and metal matrix from the atomic point of view. Furthermore, this is a good suggestion, we will try to focus on it in our future work.

In addition, the authors, when discussing the experimental data and results, should strongly highlight what are the microscopic parameters of Al and graphene justifying the results and composite behaviour. As examples:

a） what is the reason for the “rise-fall-rise” trend of cutting force in Fig. 6? The reported discussion seems too qualitative.

Response 1 a): When the FPT increases from 0.5 to 1.25 μm, the milling process goes through the minimum cutting thickness (MCT) point and the size effect point of micro-milling and finally to macro-milling condition, which results in a “rise-fall-rise” trend for milling force consistent with different milling period mechanism.

And in the revised version, we have added some data to show the extent to which milling forces increase or decrease. (Line:251-266)

When feed per tooth is 0.75μm (lower than MCT), no chip is formed and interaction between tool and workpiece is dominated by intense ploughing /rubbing force which increases with the increase of feed per tooth. The milling force reaches its peak value near FPT=0.75 um, which is 18.1%-32.6% higher than that at FPT=0.25 um.

When FPT is between 0.75 and 1μm, which is higher than the MCT, a shear-based deformation of the material happens and chips begins to form. At this time, part of the friction between tool and workpiece is converted into the shear force between the tool and the chip, which reduces the friction between tool and workpiece decreases resulting in the reduction of Fx, Fy and Fz. The milling force reaches the valley near FPT=1μm, which decreases by 10.9%-22.4% compared with that FPT=0.75μm.

With the continue increase of FPT value, chip thickness increases, the shear force between tool and chip increases while the decrease of the friction between tool and workpiece. As the FPT increased further to higher than 1μm, the transition point (size effect) from micro-milling to macro-milling reached, and the shear force plays a dominant role between tool and workpiece. In this case, the milling force rise with FPT which is consistent with that in macro-milling. In the case of FPT=1.25μm, the milling force increased by 16.1% to 31.4% compared with that FPT=1μm.

b) Fig.7: when you state "This is mainly attributed to the two factors: grain refinement effect of graphene and agglomeration of graphene.": why?

Response 1 b): Usually, the grain interface traversed by the cutter increases with the decrease of the grain size during the micromilling process which will result higher milling force.

This can be explained by the wrinkle structure of graphene which can inhibit the coarsening and growth of aluminum matrix grains and hence resulting the finer grain. In addition, the pinning or wrapping of grains also hinders grain growth during processing and finer grain are obtained.  (Line:150-163)

For 0.5% and 1.0% GNFs/Al composites with comparatively lower graphene content, Fine grain strengthening effect introduced by graphene platelets play a role (as shown in Table 1). During machining process, fine grains will hinder the shear and slip between grains and prevent the generation and propagation of fine cracks. Meanwhile, graphene can also hinder the shear and slip between grains, deflect the crack propagation in grains, and thus reduce the generation and propagation of fine cracks. All these will result comparatively higher milling force in Fx, Fy and Fz three directions. (Line:271-277)

However, a common trend observed for all mechanical properties is that there is a limiting content of GNP for enhancing properties, beyond which a remarkable switch-over is observed. The explanation for such behavior lies in the agglomeration of GNPs at high concentration, which then become stress concentrating defects. Such agglomerates promote pore formation and nucleation of micro-cracks which will result in pits on the surface of the sample after corrosion (which can be observed in the metallographic morphology in Figure 3.(Line:166-72)

For 1.5% and 2.0% GNFs/Al composites in which the content of graphene reaches a certain threshold, agglomeration effect of graphene begins to dominate, which will weaken the mechanical properties of composites and bring grain coarsening, as discussed in Table 1. This weakness and coarse grain will result lower stress required by the composite to undergo plastic deformation during milling process and eventually lead to decrease of milling force.(Line: 278-282)

c) "This is attributed to two factors: on the one hand, the wrinkle structure of graphene can inhibit the coarsening and growth of aluminum matrix grains, which resulting the finer grain and hence improvement of composites plasticity. On the other hand, since graphene is carbon particle in nature, graphene as reinforcing phase in composites have a certain lubrication effect also, thus improving the friction coefficient between tool and workpiece, and reducing the generation of friction force." These seem speculations if quantitative data are not reported. and so on.

Response 1 c): In the revised manuscript, we added grain size (measured by using Image pro plus 6.0 software) to quantify the graphene grain refinement effect in Table 1 and (Lines:136-139).

By measuring grain size with Image Pro. software, the average grain diameter of 0.5% GNFs/Al is 6.78μm, while that of 1.5% GNFs/Al and 2.0% GNFs/Al is 8.12μm and 8.37μm respectively and plain Al is 9.21μm (as shown in Table 1).

Table 1. Characteristics for different GNFs/Al

At same time, we added the friction and wear experiments to test the friction coefficients of composites with different graphene content to measure the lubrication effect of graphene on composites. Related data and statement are shown in Table 1 and Lines (180-185)

UMT-TriboLab friction and wear tester (Bruker Nano, Inc. Germany) was employed to investigate the tribology behavior of coatings in a dry sliding condition at room temperature. A Si3N4 ball with the diameter of 6 mm were selected as a counterpart. Test parameters are as follows: a normal load of 10N, sliding velocity of 5 mm/s, reciprocatory displacement of 10 mm and test time of 10 min. The worn surface was observed and examined for analyzing. The coefficient of friction (COF) obtained from the experimental results are shown in Figure 5.

Point 2: Table 1: for the experimental derived parameters, corresponding errors should be reported.

Response 2: In the revised manuscript, we add Figure 4 to show the corresponding errors of HV values. (Line: 153-154)

Point 3: Fig.9 and surface roughness: surface roughness is a very important parameter characterizing the surface morphology, topography and topology (see: A.-L. Barabasi, H.E. Stanley, Fractal Concepts in Surface Growth (Cambridge University Press, Cambridge, 1995); Nanoscale Research Letters 4, 262 (2009); Journal of Materials Processing Technology 123, 133 (2002)). Some discussions in this sense could be very useful.

Response 3: In the revised manuscript, we add related research result to the conclusions of relevant researchers to illustrate the validity of our experimental data in Lines (354-360).

At the same time, the variation trend of surface roughness value in this research is consistent with other research results. Seong et al. [36-37] established the prediction model of the minimum cutting thickness by studying the influence of friction between the tool and the workpiece on the minimum chip thickness and proposed the following formula, in which rn for tool edge radius; β for friction angle; R for tool radius; fr for feed per tooth and Ra for surface roughness. It can be obtained from the following formula that when the FPT value is ≥MCT, the value of surface roughness tends to decrease first and then increase as the FPT value increases.

; 

Point 4: Some Raman and X-ray diffraction measurement could be useful (at least as perspectives) in characterize the graphene flakes quality and the graphene-Al interaction.

Response 4: It is a good suggestion, however, due to insufficient time for us to schedule the X-ray diffraction measurement, we have only completed the Raman spectroscopy measurement of the composites. From the Raman spectra obtained, it is also sufficient for the information we need. In the revised manuscript (Lines 196-209), we add Figure 4 and related statements.

Raman spectra for different GNFs/Al composites were measured by using Bruker Senterra, as shown in Figure 6. From Figure 6, we can find that there are three main characteristic peaks for graphene. Among of them, D peak is around 1360cm-1 which indicates lattice defects of carbon atoms, G peak is around 1585cm-1 indicating the degree of graphitization, and 2D peak is around 2687cm-1 indicating the appearance and number of layers of graphene. A smaller value of ID/IG represents fewer layers of graphene flakes and better quality of graphene, which can be seen as an index for the interfacial integrity and agglomeration effect.

As can be seen from Figure 6, the increase in graphene content brings an increase of the value for both G peak and ID/IG ratio, which suggesting that the increase of graphene graphitization (graphene agglomeration) and graphene layer numbers [34]. We can also find from Figure 6 that the low D peak value for 0.5% and 1% GNFs/Al which means that the structure of graphene is comparatively complete and comparatively higher D peak values for 1.5% and 2% GNFs/Al which mainly due to the existence of a large number of microcracks, pores and lattice defects caused by agglomerated graphene through sintering process.

Author Response File: Author Response.pdf

Reviewer 2 Report

The scientific results of this work presents practical interest for the modern manufacturing industry.

The authors carried out a large complex of experimental studies. These studies are useful when embedding new types of materials such as metal matrix composites, graphene-containing materials, etc.

However, there are some comments on the manuscript text.

1. In the abstract, it is desirable to remain only specific information about the main research results presented in this paper. Other information is more relevant to the introduction.

2.There are stylistic inaccuracies, for example:

Line 17 - it is preferable to apply the term “use”

Line 19 - hardness and density cannot be "improved"; it can be either increased or decreased.

Line 21 - incorrect phrase “…the chip shape increases...” It is necessary to clarify what happens to the shape of the chip (or its morphology and size).

Line 60 — it is preferable to use the term “plasticity” instead of “ductility”.

Line 142 - the sentence is incorrect (error)

Line 143 - it is need to replace the term “ground” with “grinding”

Line 149 - it is preferable to replace “weaken” with “reduce”

Line 165-167 - insufficient explanation for the decrease in the actual density and hardness of the composite with a high graphene content. Why are the possible (and indeed observed) micropores that occur due to graphene agglomeration not taken into account? Was microhardness measured in the matrix zone and at the graphene localization sites?

Line 211-213: Each part of Figure 6 should be labeled (a), (b), etc. A text is needed to add to a figure caption for explaining of cutting conditions. A similar comment refers to Figure 7 (line 233-234).

Line 240 - what composites are meant (0.25% GNFs/Al)?

In line 263, it is preferable to use the term “core” instead of “essence”.

Line 364 - correctly "..based on the above findings"

General remarks:

It is necessary to correct some sentences in English, as the terms used in the text do not quite clearly convey the meaning of the author’s idea.

Authors are recommended correctly to place figures and tables in the text. As a rule, there is a generally accepted procedure for inserting figures in the manuscript text: first, it is placed the text with the information presented in the figure, and then the figure is placed itself (or table).

Comments for author File: Comments.pdf

Author Response

Thank you for giving us the opportunity to revise our manuscript and for the thoughtful and valuable comments on this manuscript. We have checked through the whole paper and corrected errors. At same time, we add Raman, coefficient of friction and grain size data in the revised paper. All the changes are highlighted in the Marked Revision. The point-to-point responses to the comments are listed as following.

Point 1: In the abstract, it is desirable to remain only specific information about the main research results presented in this paper. Other information is more relevant to the introduction.

Response 1: In the revised manuscript (Lines 10-23), we have checked the abstract and delete some irrelevant information and add some specific information needed.

In this paper, plain aluminum was chosen as matrix alloy and graphene reinforced aluminum alloy composites was successfully prepared via powder metallurgy approach. Micro-milling experiments were conducted to explore the effect of varying graphene nanoflakes (GNFs) content (0.5%, 1.0% and 1.5% by weight) on the machinability of composites and their machining results were compared with that of plain aluminum. Chip morphology, milling force and machined surface morphology were used as the machinability measures. Experiment results showed that when the content of GNFs is less than 1.5%, the grain refinement of GNFs plays a major role. The hardness and density of the composites are increased. When the content of GNFs is more than 1.5%, the agglomeration phenomenon is obvious, which reduces the hardness and density of the composites. Micro-milling results show that the milling force is the highest when the GNFs content is 1%, and curling degree of chips increased as FPT increase for a certain content of graphene of composites. Furthermore, when the content of GNFs in composites is more than 1%, the surface roughness of milling grooves is greatly improved, which may be related to the lubrication of graphene and the formation of continuous chips.

Point 2: It is necessary to correct some sentences in English, as the terms used in the text do not quite clearly convey the meaning of the author’s idea.

Response 2: Thanks for the suggestions, in the revised manuscript, we have checked the whole paper and made a thoroughly revision.

There are stylistic inaccuracies, for example:

Line 17 - it is preferable to apply the term “use” and Line 19 - hardness and density cannot be "improved"; it can be either increased or decreased.

Response (Line 17): The authors are grateful to the referee for pointing out these errors. In the revised version, we have correct them.

The hardness and density of the composites are increased.

Line 21 - incorrect phrase “…the chip shape increases...” It is necessary to clarify what happens to the shape of the chip (or its morphology and size).

Response (Line 19-21): In the revised version, we rewrote these statements.

Micro-milling results show that the milling force is the highest when the GNFs content is 1%, and curling degree of chips increased as FPT increase for a certain content of graphene of composites.

Line 60 — it is preferable to use the term “plasticity” instead of “ductility”.

Response (Line 54-55): In the revised version, we have corrected these statements .

Besides this, machining of MMCs is notoriously known to be difficult due to both the presence of two or more distinct phases, namely hard reinforcement phase and plastic metal matrix.

Line 142 - the sentence is incorrect (error)

Line 143 - it is need to replace the term “ground” with “grinding”; Line 149 - it is preferable to replace “weaken” with “reduce”

Response (Line 133-135) and (Line 136-142): The authors are grateful to the referees for pointing out these errors. In the revised version, we have correct them.

Before observation, the samples were grinded by using SiC paper with different meshes and then polished with 0.1 mm diamond gypsum, corroded finally by using 1% HF solution.

By measuring grain size with Image Pro. software, the average grain diameter of 0.5% GNFs/Al is 6.78μm, while that of 1.5% GNFs/Al and 2.0% GNFs/Al is 8.12μm and 8.37μm respectively and plain Al is 9.21μm (as shown in Table 1). This can be explained by the wrinkle structure of graphene which can inhibit the coarsening and growth of aluminum matrix grains and hence resulting the finer grain. In addition, the pinning or wrapping of grains also hinders grain growth during processing and finer grain are obtained.

Line 165-167 - insufficient explanation for the decrease in the actual density and hardness of the composite with a high graphene content. Why are the possible (and indeed observed) micropores that occur due to graphene agglomeration not taken into account? Was microhardness measured in the matrix zone and at the graphene localization sites?

Response (Lines 166-174): In the revised version, we have added explanation for the decrease in the decrease in the actual density and hardness of the composite with a high graphene content. And here, we explain the graphene agglomeration and micropores influence on the actual density and hardness of composites.

Besides this, in this article, microhardness was measured in the matrix zone instead of considering the graphene localization sites.

However, a common trend observed for all mechanical properties is that there is a limiting content of GNP for enhancing properties, beyond which a remarkable switch-over is observed. The explanation for such behavior lies in the agglomeration of GNPs at high concentration, which then become stress concentrating defects. Such agglomerates promote pore formation and nucleation of micro-cracks which will result in pits on the surface of the sample after corrosion (which can be observed in the metallographic morphology in Figure 3. When GNFs content in GNFs/Al composites reaches a certain threshold, 1.5% here, conglomeration of graphene (as shown in Figure 3) interrupts the consolidation and result in defects in the composite, which bring decrease hardness of GNFs/Al composite.

Line 211-213: Each part of Figure 6 should be labeled (a), (b), etc. A text is needed to add to a figure caption for explaining of cutting conditions. A similar comment refers to Figure 7 (line 233-234).

Response (Line 243-244): In the revised manuscript, we added label (a), (b), etc. for Figure 6. Figure 6 gives the milling force in three directions and resultant force under same milling conditions. So, text is not needed for explaining of cutting conditions.

And in the revised version, we deleted Figure 7 for the concise of the article since the contents of which have some repeatability with those of Figure 6.

Line 240 - what composites are meant (0.25% GNFs/Al)?

Response (Line 271-272): This is our mistake, we have correct it to 0.5% GNFs/Al in the revised version.

For 0.5% and 1.0% GNFs/Al composites with comparatively lower graphene content, Fine grain strengthening effect introduced by graphene platelets play a role (as shown in Table 1).

In line 263, it is preferable to use the term “core” instead of “essence”. Line 364 - correctly "based on the above findings"

Response (Line 292-294) and (Line 394-395): The authors are grateful to the referee for pointing out these errors. In the revised version, we have correct them .

Since the core of graphene is C particles which has a certain lubrication effect, the increase of graphene content in the composites can effectively reduce the friction between tool and workpiece.

Following conclusions can be drawn based on the above findings:

Point 3: Authors are recommended correctly to place figures and tables in the text. As a rule, there is a generally accepted procedure for inserting figures in the manuscript text: first, it is placed the text with the information presented in the figure, and then the figure is placed itself (or table).

Response 3: In the revised manuscript, we rechecked all the figures and tables and reset them as correctly as possible.

Author Response File: Author Response.pdf

Reviewer 3 Report

This is a useful tool for the practical engineering applications; I would like to suggest to the authors to make a simulation and/or other predictions for the next article as a second part of this subject. It should be better to make a detail experimental design and why not machining learning from the experimental results a detail ANOVA for giving a practical information for the industrial partners

Author Response

Thank you for giving us the opportunity to revise our manuscript and for the thoughtful and valuable comments on this manuscript. We have checked through the whole paper and corrected errors. At same time, we add Raman, coefficient of friction and grain size data in the revised paper. All the changes are highlighted in the Marked Revision. The point-to-point responses to the comments are listed as following.

Point 1: I would like to suggest to the authors to make a simulation and/or other predictions for the next article as a second part of this subject.

Response 1: Thank you for the good suggestion. The truth is in our early stage of research, we have completed part of the simulation of micro-milling process of graphene reinforced aluminum alloy composites (6065Al matrix) and obtained some results such as the cutting force, chip formation and deformation behavior of graphene during milling (as shown in the following Figure1 and Figure 2).

However, the simulation results are not comprehensive enough and still need to be further improved. So, we will discuss the milling simulation of graphene reinforced aluminum alloy composites in our future article

Point 2: It should be better to make a detail experimental design and why not machining learning from the experimental results a detail ANOVA for giving a practical information for the industrial partners.

Response 2: It is true that detailed analysis of variance would be better and more useful to ensure the primary and secondary sequence of experimental factors influence on processed surface roughness and the optimum combination of experimental factors, which is practical for the industrial partners.

In this paper, we adopt a single factor (two variables) experiment design. The main purpose of this paper here is to explore the influence of feed and graphene content on the processing performance of composite materials, and to study the influence of graphene addition on surface roughness and cutting force. The spindle speed and cutting depth are set only according to experience and research purposes. In this case, there are not sufficient experiment data for an efficient ANOVA.

Furthermore, this is a very good suggestion. We will devote ourselves to more practical experiment design for industrial production in the follow-up study.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors replied in a convincing way to all my previous comments and improved very well the manuscript.

So, now, I can support the publication of the manuscript in the present form.